Experimental validation of compressive strength prediction using machine learning algorithm

Compressive strength is one of the most important parameters for a fiber reinforced concrete. It varies with respect to the cement content, aggregate diameter and content, fiber volume fraction, length and diameter of the fibers etc. In this paper, an attempt is made to develop the soft computing model which can predict the compressive strength of the concrete if above said ingredients properties are given as the input parameters. Linear Regressor, Support Vector Regressor, LASSO regressor, Ridge Regressor, ElasticNet Regressor, Random Forest Regressor and Artificial Neural Network are the soft computing techniques used in this study to develop the prediction model. 133 data collected from the literature is used for training the model and its validation is done using the 25 data developed in the lab by conducting the compression test study. Theoretically, neural networks give the best results because they can capture non-linear relationships between features. But in our case, the results of random forest regressor are the best with minimum mean squared error. The present work uses nine essential features that affect the experiment performed to determine the strength theoretically. Thus, it aids the research community by making a comparative study of machine learning and deep learning techniques to accurately predict the compressive strength of fiber reinforced concrete. Copyright (C) 2022 Elsevier Ltd. All rights reserved.